Machine learning identifies 10 feature miRNAs for lung squamous cell carcinoma

Advances in applications of artificial intelligence algorithms for cancer-related miRNA research

MiRNAs are a class of small non-coding RNAs, which regulate gene expression post-transcriptionally by partial complementary base pairing. Aberrant miRNA expressions have been reported in tumor tissues and peripheral blood of cancer patients. In recent years, artificial intelligence algorithms such as machine learning and deep learning have been widely used in bioinformatic research. Compared to traditional bioinformatic tools, miRNA target prediction tools based on artificial intelligence algorithms have higher accuracy, and can successfully predict subcellular localization and redistribution of miRNAs to deepen our understanding. Additionally, the construction of clinical models based on artificial intelligence algorithms could significantly improve the mining efficiency of miRNA used as biomarkers. In this article, we summarize recent development of bioinformatic miRNA tools based on artificial intelligence algorithms, focusing on the potential of machine learning and deep learning in cancer-related miRNA research.

Sevoflurane attenuates proliferative and migratory activity of lung cancer cells via mediating the microRNA-100-3p/sterol O-Acyltransferase 1 axis

Recently, evidence has shown that microRNA-100-3p (miR-100-3p) has been revealed as a tumor suppressor in diverse human diseases, while its capability in lung cancer warrants further validation. In this work, we aimed to discuss the impact of sevoflurane on biological functions of lung cancer cells by modulating the miR-100-3p/sterol O-acyltransferase 1 (SOAT1) axis. Lung cancer cell lines (A549 and H460) were treated with various concentrations of sevoflurane. Cell viability, proliferation, migration, and invasion were evaluated using MTT, colony formation, wound healing, and transwell assays. Moreover, miR-100-3p and SOAT1 expressions were evaluated by reverse transcription-quantitative polymerase chain reaction in lung cancer cells. The target interaction between miR-100-3p and SOAT1 was predicted by bioinformatics analysis and verified by the dual-luciferase reporter gene assay. The findings of our work demonstrated that sevoflurane impeded the abilities on viability, proliferation, migration, and invasion of A549 and H460 cells. The expression of miR-100-3p was reduced, and SOAT1 expression was elevated in lung cancer cells. miR-100-3p targeted SOAT1. Besides, sevoflurane could lead to expressed improvement of miR-100-3p or limitation of SOAT1. Downregulation of miR-100-3p or upregulation of SOAT1 restored the suppression of sevoflurane on abilities of viability, proliferation, migration, and invasion in A549 and H460 cells. In the rescue experiment, downregulation of SOAT1 reversed the impacts of downregulation of miR-100-3p on sevoflurane on lung cancer cells. Collectively, our study provides evidence that sevoflurane restrained the proliferation and invasion in lung cancer cells by modulating the miR-100-3p/SOAT1 axis. This article provides a new idea for further study of the pathogenesis of lung cancer.

Machine-Learning-Based Late Fusion on Multi-Omics and Multi-Scale Data for Non-Small-Cell Lung Cancer Diagnosis

Differentiation between the various non-small-cell lung cancer subtypes is crucial for providing an effective treatment to the patient. For this purpose, machine learning techniques have been used in recent years over the available biological data from patients. However, in most cases this problem has been treated using a single-modality approach, not exploring the potential of the multi-scale and multi-omic nature of cancer data for the classification. In this work, we study the fusion of five multi-scale and multi-omic modalities (RNA-Seq, miRNA-Seq, whole-slide imaging, copy number variation, and DNA methylation) by using a late fusion strategy and machine learning techniques. We train an independent machine learning model for each modality and we explore the interactions and gains that can be obtained by fusing their outputs in an increasing manner, by using a novel optimization approach to compute the parameters of the late fusion. The final classification model, using all modalities, obtains an F1 score of 96.81±1.07, an AUC of 0.993±0.004, and an AUPRC of 0.980±0.016, improving those results that each independent model obtains and those presented in the literature for this problem. These obtained results show that leveraging the multi-scale and multi-omic nature of cancer data can enhance the performance of single-modality clinical decision support systems in personalized medicine, consequently improving the diagnosis of the patient.

KLF5, a Novel Therapeutic Target in Squamous Cell Carcinoma

Squamous cell carcinomas (SCCs) are the most common ectodermal cancers, and result in more than 300,000 deaths per year. The Krüppel-like family of transcription factors play a critical role in cancer pathogenesis. The Krüppel-like factor 5 gene (KLF5), which is a member of Krüppel-like family, has been reported to promote cancer cell proliferation and tumorigenesis. In this review, we discuss the roles of KLF5 in different SCCs and the mechanisms by which KLF5 transcriptionally regulates its target gene expression in the pathogenesis and progression of SCCs. Due to its significant functions in cell proliferation and differentiation, KLF5 could be a novel diagnostic biomarker and therapeutic target for the treatment of SCCs.